Earlier today I went to access a drive on my unRAID array, and it came back saying it couldn't find it (exact message varied by operating system).  PING returned replies with 0mS latency.

 

Going into the web page, on the Main tab all the drives were showing OK, but spun down ("Array Started" was the main status indication).  I clicked on one data drive and the parity drive, and they spun up.

Checked Dashboard, no errors shown, temperatures around +37C (~98F) and no SMART warnings.

Still can't get to the drives from any other machine, although Windows Explorer shows them present, and with the correct (approximately!) size and free space.

 

Oh, and clicking on "Log" in the top-right of the Main screen produces an "about:blank" window with nothing in it, and at the bottom:  "Waiting for 192.(etc)..." and just sits there like that - for half an hour and counting!

 

I'm about to try "turning it off and on again" as I'm out of ideas!

 

Anyone seen this, or know why it may be happening?

 

HP uServerl N36L, unRAID 6.1.9, running for 19+ days without problems.

Before you shut it off from the command prompt type diagnostics and then post the resulting file

 

Sent from my LG-D852 using Tapatalk

 

 

Thanks for the suggestion, however it beat you to it - it stopped responding to anything on the web page - clicking to change tabs, clicking any buttons, were all ignored.  I went to the machine and noticed that "overheated power supply" smell - anyone who has smelled it will recognise it.

 

I suspect one or more power supply capacitors may have blown (the hot weather accelerating that) and it may now have lost the 12V rail, hence not being able to spin up the disks, even to stop the array, which I tried some time ago.

 

It could be that spinning up two disks manually was all it could manage when I did that, and that finally killed the failing components.

 

So now I have some dismantling to do    I'll start by removing the disks, then power it up and check the voltages, and proceed from there.

 

Thanks again,

 

Howard

 

Thanks for the suggestion, however it beat you to it - it stopped responding to anything on the web page - clicking to change tabs, clicking any buttons, were all ignored.  I went to the machine and noticed that "overheated power supply" smell - anyone who has smelled it will recognise it.

 

I suspect one or more power supply capacitors may have blown (the hot weather accelerating that) and it may now have lost the 12V rail, hence not being able to spin up the disks, even to stop the array, which I tried some time ago.

 

It could be that spinning up two disks manually was all it could manage when I did that, and that finally killed the failing components.

 

So now I have some dismantling to do    I'll start by removing the disks, then power it up and check the voltages, and proceed from there.

 

Thanks again,

 

Howard

Hopefully not to late, but since you suspect the p/s is damaged, the absolute last thing you want to do is power it back up still connected to anything.  The 12v rail is also the main power for the CPU.  If its not completely kaput, you run the risk of damaging the mobo / cpu / gpu.  Since you had the ozone smell, just throw it out in the garbage and hope that nothing else is damaged.  Why tempt fate?

OK, seems to be good news...

 

The smell wasn't from the server, but from a dehumidifier that is on a timeswitch, and which was off when I visited the server (when I overrode the timeswitch to On, there was a rough, rumbling noise and a really strong smell from the dehumidifier!).

 

I took the drives out of the server, disconnected the motherboard power connector, and connected it to a Power Supply Tester, powered on and all voltages got green lights.  I measured the 5V and 12V on a "spare" hard drive Molex connector, and both were within 0.1V of correct.

 

So I reassembled it (you have to slide the motherboard out to get to the power connector) put the drives back in and powered it up.  unRAID came up, showing all drives, and started a Parity Check.

 

It's currently at 83% complete, with an hour to go, so it looks like whatever stopped it working before was a glitch of some sort, and turning it off and back on again was what it needed! 

 

In the meantime I've found that the PSU is a fairly standard "1U" one and there are many available as replacements, in various ratings (150W seems to have been the original).

 

So false alarm - thanks for the advice, and next time I'll try to be more sure of my facts before calling for help!

 

Cheers,

 

Howard